From mboxrd@z Thu Jan  1 00:00:00 1970
From: pieterg@gmx.com (pieterg)
Date: Thu, 8 Jul 2010 12:36:39 +0200
Subject: machine definition for colibri-pxa3xx
Message-ID: <201007081236.39550.pieterg@gmx.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

At the moment, colibri-pxa300.c as well as colibri-pxa320.c have their own 
machine definition.

However, running a kernel built with CONFIG_MACH_COLIBRI300 and 
CONFIG_MACH_COLIBRI320 on a colibri 300 or 310, it will match the 320 
machine definition, and therefore use colibri-pxa320.c.

I believe it's not possible to differentiate colibri pxa3xx based on machine 
definitions, but only by checking the their cpu.

Wouldn't it make sense to place a single "Toradex Colibri PXA3xx" machine 
definition in colibri-pxa3xx.c, and use cpu_is_pxa3nn to do cpu specific 
initialisation?
This is exactly how things are done for zylonite by the way.

I have some colibri pxa310 patches, but we need a working machine detection 
before they will actually do anything.

Rgds, Pieter